-
Notifications
You must be signed in to change notification settings - Fork 0
/
chimera.html
304 lines (295 loc) · 11.8 KB
/
chimera.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
<!DOCTYPE html>
<html lang="en">
<meta charset="UTF-8">
<link rel="icon" type="image/x-icon" href="/images/logotic.png">
<title>Project | Chimera</title>
<meta name="viewport" content="width=device-width,initial-scale=1">
<meta property="og:type" content="website">
<meta property="og:url" content="http://agalmicautomata.com/">
<meta property="og:title" content="Chimera: LP/ITX GPU List">
<meta property="og:description" content="A static list of universal and multi-compatible ITX/LP GPUs.">
<meta property="og:image" content="/images/logotic.png">
<link rel="stylesheet" href="styles.css">
<html lang="en">
</head>
<body>
<section>
<nav>
<ul>
<li><p>[Projects]</p></li>
<li><a href="index.html">[BACK]</a></li>
</ul>
</nav>
<article>
<h1>[LOW PROFILE / MODDER GPUs]</h1>
<p>This is a reference of somewhat common and modern low profile sub75W PCIe GPUs up to 2 slots tall.
The existence of this is to dedicate a page to power efficient and modular cards that you can put in almost any system. <br>
Several GPUs have been added simply as a reference, or for their notability.
<br> In the case that you need a Linux compatible GPU, the AMD GPUs listed are your best choice.
If you wish for portable drivers (so that you may install them offline, or have a back up), look into NVIDIA GPUs.
An attempt to avoid adding workstation cards was made, but because of the power of some of these cards, especially with NVIDIA,
multiple have been added. I have a webpage underconstruction about the A2000 in specific, as its a great little card with a lot of potential.<br>
As a final note, I tried to keep this mostly to "viable cards", as many low profile cards have aged past their usefulness for modern applications. See the GT710 for details.</p>
<h1>[LIST]</h1>
<style type="text/css">
.tg {border-collapse:collapse;border-color:#ccc;border-spacing:0;}
.tg td{background-color:#fff;border-color:#ccc;border-style:solid;border-width:1px;color:#333;
font-family:Arial, sans-serif;font-size:14px;overflow:hidden;padding:10px 5px;word-break:normal;}
.tg th{background-color:#f0f0f0;border-color:#ccc;border-style:solid;border-width:1px;color:#333;
font-family:Arial, sans-serif;font-size:14px;font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;}
.tg .tg-baqh{text-align:center;vertical-align:top}
.tg .tg-buh4{background-color:#f9f9f9;text-align:left;vertical-align:top}
.tg .tg-0lax{text-align:left;vertical-align:top}
.tg .tg-87db{background-color:#f9f9f9;font-size:100%;text-align:left;vertical-align:top}
</style>
<table class="tg">
<thead>
<tr>
<th class="tg-baqh">GPU</th>
<th class="tg-baqh">TDP(w)<br></th>
<th class="tg-0lax">Passmark<br>3D Score<br></th>
<th class="tg-baqh">PCIE<br>Interface</th>
<th class="tg-baqh">Cores<br></th>
<th class="tg-baqh">TMUS/<br>ROPs</th>
<th class="tg-0lax">Clock Base/<br>Boost<br></th>
<th class="tg-baqh">Memory Type/<br>Amount<br></th>
<th class="tg-baqh">Memory<br>Bus</th>
<th class="tg-baqh">Memory<br>Clock</th>
<th class="tg-baqh">Memory<br>Bandwidth<br></th>
<th class="tg-baqh">COMMENTS</th>
</tr>
</thead>
<tbody>
<tr>
<td class="tg-buh4">GT 710</td>
<td class="tg-buh4">19W</td>
<td class="tg-87db">631</td>
<td class="tg-buh4">2.0 x8</td>
<td class="tg-buh4">192</td>
<td class="tg-buh4">16/8</td>
<td class="tg-buh4">954 MHz</td>
<td class="tg-buh4">DDR3 / 2GB</td>
<td class="tg-buh4">64 bit</td>
<td class="tg-buh4">800 MHz</td>
<td class="tg-buh4">12.8 GB/s</td>
<td class="tg-buh4">Really only added this for reference to the lowest usable low profile gpu. The only usecase for this GPU is increasing iGPU headroom.<br>Used the MSI Low Profile 2GB.</td>
</tr>
<tr>
<td class="tg-0lax">RX 550</td>
<td class="tg-0lax">50W</td>
<td class="tg-0lax">2707</td>
<td class="tg-0lax">3.0 x8</td>
<td class="tg-0lax">512</td>
<td class="tg-0lax">32/16</td>
<td class="tg-0lax">1100/<br>1183 MHz</td>
<td class="tg-0lax">GDDR5 / 2GB or 4GB</td>
<td class="tg-0lax">128 bit</td>
<td class="tg-0lax">1750 MHz<br>7 Gbps</td>
<td class="tg-0lax">112 GB/s</td>
<td class="tg-0lax"></td>
</tr>
<tr>
<td class="tg-buh4">RX 460</td>
<td class="tg-buh4">75W</td>
<td class="tg-buh4">4114</td>
<td class="tg-buh4">3.0 x8</td>
<td class="tg-buh4">896</td>
<td class="tg-buh4">56/16</td>
<td class="tg-buh4">1090/<br>1200 MHz</td>
<td class="tg-buh4">GDDR5 / 2GB or 4GB</td>
<td class="tg-buh4">128 bit</td>
<td class="tg-buh4">1750 MHz<br>7 Gbps</td>
<td class="tg-buh4">112 GB/s</td>
<td class="tg-buh4"></td>
</tr>
<tr>
<td class="tg-0lax">RX 560</td>
<td class="tg-0lax">75W</td>
<td class="tg-0lax">3649</td>
<td class="tg-0lax">3.0 x8</td>
<td class="tg-0lax">1024</td>
<td class="tg-0lax">64/16</td>
<td class="tg-0lax">1175/<br>1275 MHz</td>
<td class="tg-0lax">GDDR5 / 4GB</td>
<td class="tg-0lax">128 bit</td>
<td class="tg-0lax">1750 MHz<br>7 Gbps</td>
<td class="tg-0lax">112 GB/s</td>
<td class="tg-0lax"></td>
</tr>
<tr>
<td class="tg-buh4">RX 6400<br></td>
<td class="tg-buh4">53W</td>
<td class="tg-buh4">7483</td>
<td class="tg-buh4">4.0 x4</td>
<td class="tg-buh4">768</td>
<td class="tg-buh4">48/32</td>
<td class="tg-buh4">1923/<br>2321 MHz</td>
<td class="tg-buh4">GDDR6 / 4GB</td>
<td class="tg-buh4">64 bit</td>
<td class="tg-buh4">2000 MHz<br>16 Gbps</td>
<td class="tg-buh4">128 GB/s</td>
<td class="tg-buh4"></td>
</tr>
<tr>
<td class="tg-0lax">RX 6600</td>
<td class="tg-0lax">50W</td>
<td class="tg-0lax">13608</td>
<td class="tg-0lax">4.0 x8</td>
<td class="tg-0lax">1792</td>
<td class="tg-0lax">112/64</td>
<td class="tg-0lax">989/<br>1065 MHz</td>
<td class="tg-0lax">GDDR6 / 8GB</td>
<td class="tg-0lax">128 bit</td>
<td class="tg-0lax">???? MHz<br>13.2 Gbps</td>
<td class="tg-0lax">224 GB/s</td>
<td class="tg-0lax">An embedded application version of the RX6600 full size, by a FSI Embedded under the product code EP23GL-LP4AU. Uses mini displayports like most enterprise GPUs of the same use. Added due to the odd similarities to the A2000.</td>
</tr>
<tr>
<td class="tg-buh4">ARC A310</td>
<td class="tg-buh4">30W</td>
<td class="tg-buh4">4099</td>
<td class="tg-buh4">4.0 x8</td>
<td class="tg-buh4">768</td>
<td class="tg-buh4">32/16</td>
<td class="tg-buh4">1750/<br>1750 MHz</td>
<td class="tg-buh4">GDDR6 / 4GB</td>
<td class="tg-buh4">64 bit</td>
<td class="tg-buh4">1937 MHz<br>15.5 Gbps</td>
<td class="tg-buh4">124 GB/s</td>
<td class="tg-buh4">Resizable bar is almost required to get full performance, but works without.</td>
</tr>
<tr>
<td class="tg-0lax">ARC A380</td>
<td class="tg-0lax">75W</td>
<td class="tg-0lax">5088</td>
<td class="tg-0lax">4.0 x8</td>
<td class="tg-0lax">1024</td>
<td class="tg-0lax">64/32</td>
<td class="tg-0lax">2000/<br>2050 MHz</td>
<td class="tg-0lax">GDDR6 / 6GB</td>
<td class="tg-0lax">96 bit</td>
<td class="tg-0lax">1937 MHz<br>15.5 Gbps</td>
<td class="tg-0lax">186 GB/s</td>
<td class="tg-0lax">Resizable bar is almost required to get full performance, but works without.</td>
</tr>
<tr>
<td class="tg-buh4">GTX 1630</td>
<td class="tg-buh4">75W</td>
<td class="tg-buh4">4998</td>
<td class="tg-buh4">3.0 x16</td>
<td class="tg-buh4">512</td>
<td class="tg-buh4">32/16</td>
<td class="tg-buh4">1740/<br>1785 MHz</td>
<td class="tg-buh4">GDDR6 / 4GB</td>
<td class="tg-buh4">64 bit</td>
<td class="tg-buh4">1500 MHz<br>12 Gbps</td>
<td class="tg-buh4">96 GB/s</td>
<td class="tg-buh4"></td>
</tr>
<tr>
<td class="tg-0lax">GTX 1650<br></td>
<td class="tg-0lax">75W</td>
<td class="tg-0lax">7872</td>
<td class="tg-0lax">3.0 x16</td>
<td class="tg-0lax">896</td>
<td class="tg-0lax">56/32</td>
<td class="tg-0lax">1285/<br>1665 MHz</td>
<td class="tg-0lax">GDDR5 / 4GB</td>
<td class="tg-0lax">128 bit</td>
<td class="tg-0lax">2001 MHz<br>8 Gbps</td>
<td class="tg-0lax">128.1 GB/s</td>
<td class="tg-0lax"></td>
</tr>
<tr>
<td class="tg-buh4">RTX 3050 LP</td>
<td class="tg-buh4">70W<br>(80W)*</td>
<td class="tg-buh4">12171</td>
<td class="tg-buh4">4.0 x8</td>
<td class="tg-buh4">2304</td>
<td class="tg-buh4">72/32</td>
<td class="tg-buh4">1042/<br>1470 MHz</td>
<td class="tg-buh4">GDDR6 / 6GB</td>
<td class="tg-buh4">96 bit</td>
<td class="tg-buh4">1750 MHz<br>14 Gbps</td>
<td class="tg-buh4">168 GB/s</td>
<td class="tg-buh4">*While most PCIE slots should be able to handle 80w, it should be kept in mind that peak draw will spike past the 75w standard.</td>
</tr>
<tr>
<td class="tg-0lax">RTX 4060 LP</td>
<td class="tg-0lax">115W</td>
<td class="tg-0lax">19698</td>
<td class="tg-0lax">4.0 x8</td>
<td class="tg-0lax">3072</td>
<td class="tg-0lax">96/48</td>
<td class="tg-0lax">2460/<br>2475 MHz</td>
<td class="tg-0lax">GDDR6 / 8GB</td>
<td class="tg-0lax">128 bit</td>
<td class="tg-0lax">2125 MHz<br>17 Gbps</td>
<td class="tg-0lax">272 GB/s</td>
<td class="tg-0lax">The usage of an 8 pin connector is required, and halves the use cases of this GPU. I genuinely have no idea what they were thinking when doing this, and there was no reason to not tune this card to use a 6 pin or no connector.</td>
</tr>
<tr>
<td class="tg-buh4">RTX A2000 6 GB</td>
<td class="tg-buh4">69W</td>
<td class="tg-buh4">13782</td>
<td class="tg-buh4">4.0 x16</td>
<td class="tg-buh4">3328</td>
<td class="tg-buh4">104/48</td>
<td class="tg-buh4">562/<br>1200 MHz</td>
<td class="tg-buh4">GDDR6 / 6GB</td>
<td class="tg-buh4">192 bit</td>
<td class="tg-buh4">1500 MHz<br>12 Gbps</td>
<td class="tg-buh4">288 GB/s</td>
<td class="tg-buh4"></td>
</tr>
<tr>
<td class="tg-0lax">RTX A2000 12 GB</td>
<td class="tg-0lax">69W</td>
<td class="tg-0lax">13696</td>
<td class="tg-0lax">4.0 x16</td>
<td class="tg-0lax">3328</td>
<td class="tg-0lax">104/48</td>
<td class="tg-0lax">562/<br>1200 MHz</td>
<td class="tg-0lax">GDDR6 / 12GB</td>
<td class="tg-0lax">192 bit</td>
<td class="tg-0lax">1500 MHz<br>12 Gbps</td>
<td class="tg-0lax">288 GB/s</td>
<td class="tg-0lax"></td>
</tr>
<tr>
<td class="tg-buh4">ADA A4000 SFF</td>
<td class="tg-buh4">70W</td>
<td class="tg-buh4">24269</td>
<td class="tg-buh4">4.0 x16</td>
<td class="tg-buh4">6144</td>
<td class="tg-buh4">192/80</td>
<td class="tg-buh4">720/<br>1560 MHz</td>
<td class="tg-buh4">GDDR6 / 20GB</td>
<td class="tg-buh4">160 bit</td>
<td class="tg-buh4">1750 MHz<br>14 Gbps</td>
<td class="tg-buh4">280 GB/s</td>
<td class="tg-buh4">This thing is unreasonably powerful, and exactly what the 4060LP should have been.</td>
</tr>
<tr>
<td class="tg-0lax">RTX A4000 ITX<br>(Shunt Modded)</td>
<td class="tg-0lax">205W<br>(1.82x)</td>
<td class="tg-0lax">19321</td>
<td class="tg-0lax">4.0 x16</td>
<td class="tg-0lax">6144</td>
<td class="tg-0lax">192/96</td>
<td class="tg-0lax">735/<br>2205 MHz</td>
<td class="tg-0lax">GDDR6 / 16GB</td>
<td class="tg-0lax">256 bit</td>
<td class="tg-0lax">2000 MHz</td>
<td class="tg-0lax">448 GB/s</td>
<td class="tg-0lax">This is not low profile, nor is it slot powered. But I had to add this simply for how powerful this card is for its size. Created by JK&G Ventures, it uses a 3060 ITX cooler, 6 to 8 pin replacement and a shunt mod.</td>
</tr>
</tbody>
</table>
</article>
</section>
<footer>
<p>AgalmicAutomata, held by Aternalogical, dedicated to open source and self sufficient designs.</p>
</footer>
</body>
</html>